What problem does it solve? Skills accumulate value only when real session learnings are written back into them; without a structured process, debugging discoveries, data quirks, and workflow fixes are lost when the session ends. ## Core Features & Use Cases - Session Learning Extraction: Scans the conversation for new error patterns, data and schema surprises, query techniques, and configuration quirks worth capturing. - Lean Skill Updates: Reads each target SKILL.md first, avoids duplication, and appends minimal bullets, short code snippets, or essential doc references to the right section. - Approval-Gated Edits: Presents a summary of proposed updates per skill and applies changes only after user approval, then checks workflow.md cross-references. - Use Case: After a long dlt debugging session where you discovered a pagination quirk and a useful ibis query pattern, run this Skill to fold those findings into the relevant skills so the next session starts smarter. ## Quick Start Ask the assistant to review this session and update the relevant skills with anything new we learned.